Full text of DOI:10.1093/gerona/55.11.B533

Journal of Gerontology: BIOLOGICAL SCIENCES
Copyright 2000 by The Gerontological Society of America
2000, Vol. 55A, No. 11, B533–B536
Telomere Shortening With Aging in Human Liver
Kaiyo Takubo,1,2,3 Ken-Ichi Nakamura,1 Naotaka Izumiyama,1 Eiki Furugori,1 Motoji Sawabe,1,2
Tomio Arai,1,2 Yukiyoshi Esaki,1,2 Ken-Ichi Mafune,4 Makoto Kammori,5 Mutsunori Fujiwara,6
Motonobu Kato,7 Mitsuo Oshimura,7 and Koji Sasajima3
1Department of Clinical Pathology, Tokyo Metropolitan Institute of Gerontology, Japan.
2Department of Pathology, Tokyo Metropolitan Geriatric Hospital, Japan.
3First Department of Surgery, Nippon Medical School, Japan.
4Department of Surgery II and 5Department of Surgery III, University of Tokyo Faculty of Medicine, Japan.
6Department of Pathology, Japanese Red Cross Medical Center, Tokyo, Japan.
7Department of Molecular and Cell Genetics, School of Life Sciences, Tottori, Japan.
Progressive telomere shortening with aging was studied in the normal liver tissue of 94 human subjects aged be-
tween 0 and 101 years old to determine the rate of telomere loss in 1 year. Telomere length demonstrated age-
related shortening with reduction of 55 base pairs (bp) per year. The mean telomere length in five neonates was
12.9 Ϯ 2.6 kilobase pairs (kbp), and that in one centenarian was 8.3 kbp. Mean telomere lengths by age group
were 13.2 Ϯ 2.0 kbp (Յ8 years; 10 subjects), 7.8 Ϯ 1.9 kbp (40–79 years; 29 subjects), and 7.5 Ϯ 2.0 kbp (Ն80
years; 53 subjects), with reduction thus appearing to show slowing on the attainment of middle age. The differ-
ence of mean telomere lengths for two groups with or without advanced malignancies of other than liver origin
was not significant in the older two groups. Despite the slow turnover of liver tissue, the overall reduction rate of
telomere length decrease in 1 year was almost the same as that of digestive tract mucosa, with its very rapid renewal.
ORMAL human cells exhibit a limited capacity to
proliferate in culture (1), and this finite replicative life
MATERIALS AND METHODS
Samples of normal hepatic tissue were obtained from the
left lobe of the liver from 94 individuals without metastatic
carcinoma, hepatocellular carcinoma, liver cirrhosis, or hep-
atitis (46 males and 48 females between 0 and 101 years of
age), including five neonates (0–4 weeks), two children (1–
2 years) and one centenarian (more than 99 years of age),
who underwent autopsy within 5 hours after death at the To-
kyo Metropolitan Geriatric Hospital or the Japanese Red
Cross Medical Center. Only two subjects could be exam-
ined who were in their 20s and 30s. The causes of death in
the five neonates were congenital obstruction of the small
intestine, congenital abdominal wall defect, sepsis, holo-
prosencephaly, and Bochdalek’s hernia. The causes of death
in the two children were cardiomyopathy and transposition
of the great vessels. All samples were stored at Ϫ80ЊC until
use. The study protocol was approved by the Tokyo Metro-
politan Institute of Gerontology Ethical Committee.
Preliminarily, for five of the 94 subjects, hepatocytes
were isolated from hepatic tissue by collagenase perfusion
through the portal vein as previously described (12). Briefly,
fresh tissue blocks from autopsied livers were perfused with
collagenase from the portal veins, and total liver cell sus-
pensions were centrifuged at 50 g for 1 minute. The super-
natant was further centrifuged at 50 g for 3 minutes to pellet
hepatocytes.
N
span is frequently used as a model of human aging. The
phenomenon is considered to be associated with reduction
in telomere length as an indicator of the number of cell divi-
sions undergone (2,3). Human fibroblasts in culture usually
stop proliferating with a telomere length of approximately
five to six kilobase pairs (kbp); this is known as cellular se-
nescence (4,5). Telomeres are located at the ends of eukary-
otic chromosomes, where they are considered to protect
against degeneration, reconstruction, fusion, and loss (2)
and to ensure complete replication of DNA molecules (6).
Telomere DNA is very simple in structure, comprising hun-
dreds to thousands of TTAGGG repeats in humans (7) and
other vertebrates. Although it has been known for more than
10 years that these repeats are shortened by 50–150 base
pairs (bp) at each cell division in human fibroblasts and
lymphocytes in vitro (8,9), telomere reduction with aging in
humans has not been clarified for most tissues. We have
therefore conducted systemic studies to assess the telomere
length from human tissues of all types (10,11), including the
liver. To our knowledge, there has hitherto been no system-
atic measurement of telomere length in the normal human
liver. Southern blot analysis has revealed large differences
among individuals (11), so many subjects must be examined
for optimal results. For the present study we measured
terminal restriction fragments (TRFs), containing tandem
repeats of TTAGGG, by Southern blotting in normal liver
tissue obtained from 94 autopsied patients with no meta-
static carcinoma, hepatocellular carcinoma, liver cirrhosis,
or hepatitis.
From the whole liver tissues from the 94 subjects and
from the isolated hepatocytes from the five subjects, ge-
nomic DNA was prepared by using proteinase K and sodium
dodecyl sulfate (SDS) followed by repeated phenol-chloro-
form extractions. Five-microgram aliquots were digested
B533

B534
TAKUBO ET AL.
with the restriction enzyme Hinf I (Boehringer Mannheim
Biochemica, Germany), with complete cutting being con-
firmed by electrophoresis of the DNA digests on 0.8% aga-
rose gels. Fractionated DNA fragments were transferred to
nylon membranes (Hybond-Nϩ, Amersham, UK) by an al-
kaline transfer technique using capillary blotting, followed
by hybridization for 12 hours at 50ЊC in an appropriate solu-
tion [6ϫ SSPE (1ϫ; 0.15 M of NaCl, 10 mM of sodium
phosphate, and 1 mM of ethylenediamine tetra-acetic acid;
pH 7.4), 1% SDS] to a (TTAGGG)4 probe labeled with [␥-
32P]ATP (where ATP is adenosine triphosphate; Amer-
sham) at the 5Ј end using T4 polynucleotide kinase (Toy-
obo, Japan). The membranes were washed in 2ϫ SSC
(17.55 g/l of NaCl, and 8.82 g/l of sodium citrate) at room
temperature, and then in 6ϫ SSC, 0.1% SDS at 50ЊC for 15
minutes while being shaken. They were then dried with fil-
ter papers and exposed to Fuji Imaging Plates (Fuji Photo
Film Co. Ltd., Japan) for 3 hours at room temperature.
Analysis was with a BAS-2500 Mac image analyzer (Fuji
Photo Film), with the use of the programs Image Reader
(Version 1.1, Fuji Photo Film) and Mac Bas (Version 2.4,
Fuji Photo Film). TRF lengths were determined by compar-
ing the position of the maximum radioactivity in each lane
with molecular size markers. Differences in mean values
were assessed for significance with the Student’s t test and
correlations by Fisher’s test.
Figure 2. Southern blot analysis of telomeres of human hepatic tis-
sue from subjects aged from 0 to more than 100 years old. Sizes are
indicated on the left. Radiograms for samples from 1-day-old individ-
uals (two subjects) and individuals aged 2, 5, 33, 57, 68 (two subjects),
95, and 101 years are shown. Terminal restriction fragment lengths
are, respectively, 17.1, 13.2, 14.9, 13.6, 9.2, 10.6, 8.6, 9.9, 6.4, and 8.3 kbp.
RESULTS
In the preliminary experiment, the differences in telomere
lengths between whole tissue samples and isolated hepato-
cytes for the five subjects were from Ϫ0.2 to ϩ0.6 kbp,
without significant difference (Figure 1). Accordingly, we
regarded telomere lengths of whole liver tissues as those of
hepatocytes. The smear widths on Southern blot analysis of
telomeres of hepatic tissue increased with age (Figure 2).
Figure 3 summarizes telomere lengths for all 94 subjects,
plotted as a function of subjects’ age. According to the
Fisher’s test in all 94 subjects, age-related shortening was
highly significant (r ϭ Ϫ.6, p Ͻ .0001), with average telo-
mere length shortening by 55 bp per year. With the same
test in the 82 subjects more than 40 years old, average telo-
mere length shortening by 12 bp per year was recognizable,
but this was not significant (data not shown).
Mean telomere length in five neonates was 12.9 Ϯ 2.6
kbp, and the value for one centenarian was 8.3 kbp. Mean
Figure 1. Southern blot analysis of telomeres of the whole liver tis-
sues and isolated hepatocytes from two 83-year-old men (left), and
imaging scans of these telomere hybridization signals (middle and
right). Lanes 1 and 3 show whole liver tissues and lanes 2 and 4 show
isolated hepatocytes. Telomere lengths of lanes 1 and 2 are 7.0 and
7.2 kbp, and those of lanes 3 and 4 are 7.5 and 6.9 kbp, respectively.
Figure 3. Loss of telomere DNA in human hepatic tissue as a func-
tion of subject age. Mean terminal restriction fragment (TRF) lengths
from a quantitative analysis of autoradiograms, as described in the
Materials and Methods section, were plotted as a function of subject’s
age. The slope (Ϫ55 bp/y) of the linear regression line is significantly
different from zero ( p Ͻ .0001).

TELOMERE SHORTENING IN HUMAN LIVER
B535
telomere lengths for three age groups were 13.2 Ϯ 2.0 kbp
(Յ8 years; 10 subjects), 7.8 Ϯ 1.9 (40–79 years; 29 sub-
jects), and 7.5 Ϯ 2.0 kbp (Ն80 years; 53 subjects). The dif-
ference between the youngest and oldest groups was signifi-
cant ( p Ͻ .0001). However, this was not the case for the
oldest two age groups.
Mean telomere lengths for two groups, those with sub-
jects 40–79 years old with (17 subjects) or without (12 sub-
jects) advanced malignancies of other than liver origin,
were 8.0 Ϯ 2.2 and 7.6 Ϯ 5.7 kbp; difference was not sig-
nificant. Mean telomere lengths for two groups with sub-
jects above 80 years of age (17 subjects) or without (36 sub-
jects) advanced malignancies were 7.3 Ϯ 4.1 and 7.6 Ϯ 4.1
kbp, again without significant difference.
Subjects with telomere lengths more than 14 kbp were
observed in 30.0% (Յ8 years), 0% (40–79 years), and 0%
(Ն80 years). Telomere lengths of 11–14 kbp were observed
in 60.0% of those aged Յ8 years, 6.9% of those aged 40–79
years, and 5.7% of those Ն80 years. Respective figures for
lengths of 8–11 kbp were 10.0%, 37.9%, and 30.2%, and
those for less than 8 kbp were 0%, 55.2%, and 64.2% (Fig-
ure 4). The two subjects aged 9–39 years were excluded
from Figure 4.
a renewal time for human hepatocytes has not been re-
ported, that for mouse hepatocytes was found to be 480–620
days (1.3–1.7 years) in an autoradiographic study (16).
Telomere repeats are shortened by 50–150 bp at each cell
division in human fibroblasts in vitro and lymphocytes in
vitro (4,5,8). In contrast, turnover of esophageal and colonic
mucosal epithelia is very rapid at approximately 3–7 days
(17,18), and we previously revealed reduction of only 60
and 67 bp respectively in 1 year in these tissues (10,11). Al-
though the turnover time is very different among the tissues,
annually telomere reduction rates among them are not very
different. Thus, the rate of telomere shortening of hepatic
tissue in 1 year is almost the same as those for esophageal
and colonic mucosae, suggesting that germ cells in the latter
have a mechanism for lengthening telomeres.
As to the telomere length of the hepatic tissue, hepatocel-
lular carcinomas demonstrated a shorter telomere length
than the corresponding noncancerous liver tissues in 20
cases, with respective mean values of 5.4 and 8.8 kbp (19).
Although many different research groups (20,21) have also
measured the telomere lengths of hepatocellular carcinomas
and corresponding noncancerous liver tissues, the noncan-
cerous tissues were not in a normal condition but rather
demonstrated chronic hepatitis or liver cirrhosis. There has
been no systematic study of telomere length in normal hu-
man liver. Kojima and colleagues (20) have reported the
mean telomere length in eight normal livers (43–68 years)
to be 7.3 Ϯ 1.4 kbp, and Miura and coworkers (21) de-
scribed a value for four normal livers (13–79 years) of
8.9 Ϯ 1.2 kbp. Urabe and associates (22) have reported
telomere lengths for 13 normal autopsied livers (one was
from a 14-year-old individual, and the remaining 12 were
from individuals aged more than 40 years; the mean age
was 64.1 years) but could not demonstrate a significant de-
crease with aging. According to Fisher’s test, in the 82 sub-
jects more than 40 years old, age-related shortening was not
significant. Moreover, there was no significant difference
between the oldest two age groups (40–79 years and Ն80
years).
DISCUSSION
Telomeres from all human tissues are generally consid-
ered to shorten on cell division, except in carcinomatous
and germ cells (2,3,6,8), although this has not been con-
firmed for many cases. Reduction has been reported for pe-
ripheral lymphocytes in vivo (41 bp/y, ages 0–107 years)
(9), skin fibroblasts in vitro (15 bp/y, ages 0–93 years) (4),
epidermal cells in vivo (19.8 bp/y, ages 0–92 years) (13),
peripheral blood cells in vivo (33 bp/y) (8), human mucosae
from large and small intestines (42 bp/y, ages 0–89 years)
(14), human large bowel mucosa from surgical and autopsy
specimens (67 bp/y, ages 0–97 years) (10), esophageal mu-
cosa (60 bp/y, ages 0–102 years) (11), and vascular intimal
tissues (15). Here we obtained data pointing to an average
reduction rate in human hepatic tissue of 55 bp/y. Although
It has occasionally been reported that rapid telomere loss
occurs in the young, and only slow reduction is typical in
the elderly. Thus Frenck and colleagues (23) reported that
rapid loss of telomeres from peripheral blood cells in the
young contrasted with a much slower rate of loss in older
adults. Esophageal mucosa also has been suggested to ex-
hibit rapid reduction of telomere length in the young (11).
These data suggest the possibility that rapid shortening oc-
curs in the young and only a slower reduction occurs after
40 years of age, not only in the liver but also in many other
human organs or tissues. However, in human fibroblasts no
rapid reduction has been reported (4), so tissue or cell-spe-
cific differences might also be expected.
Our results indicate that telomere length in the hepatic
tissue is not influenced by malignant disease originating in
other organs.
Telomerase is a ribonucleoprotein that synthesizes telo-
mere DNA onto chromosomal ends by using a segment of
its RNA component as a template (24). Strong telomerase
activity has been detected in human germline cells and car-
cinoma tissues, and in established cell lines in culture (25).
Figure 4. Distribution of telomere lengths in three different age
groups. The two subjects aged 9–39 years were excluded (TRF; termi-
nal restriction fragment).

B536
TAKUBO ET AL.
10. Nakamura K, Furugori E, Takubo K, et al. Telomere length in normal
mucosa and cancer of the digestive bowel. Tumor Markers (Shuyo
Makers). 1999;14:33–35 (in Japanese).
11. Takubo K, Nakamura K, Izumiyama N, et al. Telomere shortening
with aging in human esophageal mucosa. Age. 1999;22:95–99.
12. Senoo H, Hata R. Isolation of perisinusoidal stellate cells (Vitamin
A-storing cells, fat-storing cells) of the rat liver [in Japanese]. Connect
Tissue. 1993;25:129–137.
13. Lindsey J, McGill NI, Lindsey LA, Green DK, Cooke HJ. In vivo loss
of telomeric repeats with age in humans. Mutat Res. 1991;256:45–48.
14. Hiyama E, Hiyama K, Tatsumoto N, Kodama T, Shay JW, Yokoyama
T. Telomerase activity in human intestine. Int J Oncol. 1996;9:453–458.
15. Chang E, Harley CB. Telomere length and replicative aging in human
vascular tissues. Proc Natl Acad Sci USA. 1995;92:11190–11194.
16. Cameron IL. Cell renewal in the organs and tissues of the nongrowing
adult mouse. Texas Rep Biol Med. 1970;28:203–248.
17. Eastwood GL. Gastrointestinal epithelial renewal. Gastroenterology.
1977;72:962–975.
18. Marques-Pereira JP, Leblond CP. Mitosis and differentiation in the
stratified squamous epithelium of the rat esophagus. Am J Anat. 1965;
117:73–90.
19. Ohashi K, Tsutsumi M, Kobitsu K, et al. Shortened telomere length in
hepatocellular carcinomas and corresponding background liver tissues
of patients infected with hepatitis virus. Jpn J Cancer Res. 1996;87:
419–422.
20. Kojima H, Yokosuka O, Imazeki F, Saisho H, Omata M. Telomerase
activity and telomere length in hepatocellular carcinoma and chronic
liver disease. Gastroenterology. 1997;112:493–500.
In contrast, it appears to be lacking in most normal human
somatic cells. Keratinocytes of the skin epidermis (26) and
endothelial cells (27) in culture have been reported to dem-
onstrate weak telomerase activity. Moreover, it has been re-
cently shown that nonneoplastic peripheral leukocytes (28),
gastric mucosa (29), esophageal mucosa (30,31), and co-
lonic mucosa (14) have detectable telomerase activity.
However, there has been no such report for hepatic tissue.
Although human liver tissue consists of hepatocytes, si-
nusoidal and vessel endothelial cells, Kupffer stellate cells,
Ito cells, bile duct and capillary cells, and inflammatory
cells, hepatocytes constitute 79% of the volume (32). In an-
other report hepatocytes accounted for 78% of hepatic tis-
sue volume (33). Ogawa and colleagues (34) described the
relative numbers of these four cellular components in he-
patic tissue to be 100 for hepatocytes, 27.5 for endothelial
cells, 15.8 for Kupffer cells, and 10.1 for Ito cells. Accord-
ingly, approximately 64% of all cells in the liver are hepato-
cytes. Our data for telomere length in the hepatic tissue
point to a mean reduction rate of 55 bp/y for all these cells
mixed together. However, our preliminary experiment com-
paring the telomere lengths between the whole liver tissues
and isolated hepatocytes demonstrated very similar values.
We therefore conclude that telomere lengths of mixed cells
can be regarded to reflect those of hepatocytes themselves.
In the present study, telomere shortening predominantly
occurred before 40 years of age, as in previous investigation
of leukocytes (23) and esophageal mucosa (11). The reason
for this age-dependent shortening requires explanation, but
presumably the slowing down with age may be related to a
gradual reduction in cell turnover with advancing years.
21. Miura N, Horikawa I, Nishimoto A, et al. Progressive telomere short-
ening and telomerase reactivation during hepatocellular carcinogenesis.
Cancer Genet Cytogenet. 1997;93:56–62.
22. Urabe Y, Nouso K, Higashi T, et al. Telomere length in human liver
diseases. Liver. 1996;16:293–297.
23. Frenck RW Jr, Blackburn EH, Shannon KM. The rate of telomere se-
quence loss in human leukocytes varies with age. Proc Natl Acad Sci
USA. 1998;95:5607–5610.
24. Greider CW, Blackburn EH. Identification of a specific telomere ter-
minal transferase activity in Tetrahymena extracts. Cell. 1985;43:405–
413.
25. Kim NW, Piatyszek MA, Prowse KR, et al. Specific association of hu-
man telomerase activity with immortal cells and cancer. Science.
1994;266:2011–2015.
26. Yasumoto S, Kunimura C, Kikuchi K, et al. Telomerase activity in
normal human epithelial cells. Oncogene. 1996;13:433–439.
27. Hsiao R, Sharma HW, Ramakrishnan S, Keith EY, Narayanan R.
Telomerase activity in normal human endothelial cells. Anticancer
Res. 1997;17:827–832.
Acknowledgments
This work was partly supported by a Grant-in-Aid (7-7) for Cancer Re-
search from the Ministry of Health and Welfare and a Grant-in-Aid (09470264)
from the Ministry of Education, Science, Sports and Culture of Japan.
Address correspondence to Kaiyo Takubo, MD, Department of Clinical
Pathology, Tokyo Metropolitan Institute of Gerontology, Sakaecho,
Itabashi-ku, Tokyo 173-0015, Japan. E-mail: takubo@tmig.or.jp
28. Counter CM, Gupta J, Harley CB, Leber B, Bacchetti S. Telomerase
activity in normal leukocytes and in hematologic malignancies. Blood.
1995;85:2315–2320.
References
29. Hiyama E, Yokoyama T, Tatsumoto N, et al. Telomerase activity in
gastric cancer. Cancer Res. 1995;55:3258–3262.
30. Takubo K, Nakamura K, Izumiyama N, et al. Telomerase activity in
esophageal carcinoma. J Surg Oncol. 1997;66:88–92.
1. Hayflick L. The limited in vitro lifetime of human diploid cell strains.
Exp Cell Res. 1965;37:614–636.
2. Blackburn EH. Structure and function of telomeres. Nature. 1991;350:
569–573.
31. Morales CP, Lee EL, Shay JW. In situ hybridization for the detection
of telomerase RNA in the progression from Barrett’s esophagus to
esophageal adenocarcinoma. Cancer. 1998;83:652–659.
32. Rohr HP, Lüethy J, Gudat F, et al. Stereology: A new supplement to the
study of human biopsy specimens. In: Popper H, Schaffner F, ed. Progress
in Liver Diseases. New York: Grune and Stratton; 1976;5:24–34.
33. Blouin A, Bolender RP, Weibel ER. Distribution of organelles and
membranes between hepatocytes and nonhepatocytes in the rat liver
parenchyma. J Cell Biol. 1977;72:441–455.
3. Greider CW. Telomere length regulation. Annu Rev Biochem. 1996;
65:337–365.
4. Allsopp RC, Vaziri H, Patterson C, et al. Telomere length predicts rep-
licative capacity of human fibroblasts. Proc Natl Acad Sci USA 1992;
89:10114–10118.
5. Harley CB, Futcher AB, Greider CW. Telomeres shorten during age-
ing of human fibroblasts. Nature. 1990;345:458–460.
6. Levy MZ, Allsopp RC, Futcher AB, Greider CW, Harley CB. Telomere
end-replication problem and cell aging. J Mol Biol. 1992;225:951–960.
7. Moyzis RK, Buckingham JM, Cram LS, et al. A highly conserved re-
petitive DNA sequence, (TTAGGG)n, present at the telomeres of
human chromosomes. Proc Natl Acad Sci USA. 1988;85:6622–6626.
8. Hastie ND, Dempster M, Dunlop MG, Thompson AM, Green DK,
Allshire RC. Telomere reduction in human colorectal carcinoma and
with ageing. Nature. 1990;346:866–868.
34. Ogawa K, Minase T, Yokoyama S, Onoe T. An ultrastructural study of
peroxidatic and phagocytic activities of two types of sinusoidal lining
cells in rat liver. Tohoku J Exp Med. 1973;111:253–269.
9. Vaziri H, Schächter F, Uchida I, et al. Loss of telomeric DNA during
aging of normal and trisomy 21 human lymphocytes. Am J Hum
Genet. 1993;52:661–667.
Received June 14, 1999
Accepted March 28, 2000
Decision Editor: Jay Roberts, MD